reflection dynamics

  • Overview

    'Rocket Attack' – the name of which is not designed very well – is a skill-based game where the player needs to defeat enemies to gain further access to the game. The game is separated into two main phases: the buying phase and the combat phase. The goal of the game is to maintain an army during the buying phase and kill as much enemy units as possible during the combat phase. These enemies will provide money to buy and repair new units that will aid the player in future fights. As the player progresses, these enemy units will increase in strength. In order to make it to the end of the game, the player has to balance his acts between the buying phase and the combat phase.

    What went well?

    Designing the maintenance phase of the game (which is a fancy name for the 'buying of units' phase) went pretty smoothly. First, I wanted to add only drones, which attack enemy units but when I prototyped this, it made not as much sense because that lacked a real dynamic economy. The next thing I came up with were banks, which provide a steady money flow for the player but they are also quite vulnerable and cannot move. This adds much more to the dynamic of the economy because it gives the player a choice what to build and base his future strategy on it. After that, the choice to add mines was simple because the implementation wasn't very difficult and it makes room for more strategy, although it doesn't add much to the dynamic economy.

    What went wrong?

    Adding the dynamic economy was really hard to balance because it is a very delicate design. If the player made too much money after each wave, the game wasn't any fun because the player could just let his army fight all the waves. If he got too little money, the game wouldn't be fun either because the player wouldn't be able to buy anything and die all the time. To find the right balance for the mines, drones and enemies turned out to be a lot harder as well, for approximately the same reasons as balancing the money flow. During play-tests with different people (some casual, like my roommates, some hardcore, like people from school) I had to adjust the values again and again.

  • Feedback Loops

    There are three feedback loops that have the most noticeable influence on the game. Two of those are negative and one is positive.

    One, Positive
    As the player kills enemies he gets bounty for those enemies. With that money he can buy more mines, drones and banks. Mines will make more money than drones but are also more vulnerable and are unable to move, while drones and mines (also stationary) will help the player kill more enemies and with that get more bounty. With this extra money, the player can build more structures, making the loop complete.

    Two, Negative
    As the player builds more structures, the enemies are more likely to hit those structures and destroy them. A structure will slowly regenerate health at the cost of money. The more structures are hit, the more money will be used to repair those structures and when a structure is destroyed the earlier mentioned positive feedback loop will lose some of its acceleration as well.

    Three, Negative
    When a structure is destroyed, it will go down in an explosion, damaging structures around it, which will then get damaged causing the player to lose additional money for repairs and the chance to create a chain reaction of structures that will get destroyed, again reducing the acceleration of the first, positive feedback loop.

  • Economy Building

    there are three economy building mechanisms in Rocket Attack: One
    The player gets money from killing enemies (bounty) he can then spent this money on drones, mines and banks, which will gather additional money for the player.

    The player needs to take care of his structures, because enemies try to destroy them.

    The player's health is the final economy, because he needs to protect his structures but he mustn't die in the process.


    During play-tests in the dynamics theme of the game, the balance was very important. It was difficult to make the game challenging but not impossible and have the Building/ Maintaining phase of the game add just enough, without the enemies or structures being killed instantly leaving little window open for interactive and fun play. This was hard and had to be tested again and again.

    The second thing that I got commented on was that adding a bounty system wasn't really that useful if there was a fixed number of enemies in every wave (because no matter how well the player performed, he would always get the same amount of money from bounties). I agreed to this statement and changed the enemy waves in such a way that there was a fixed time window in which enemies would spawn. It was up to the players performance how much enemies would be killed in those time windows.